/external/libvpx/vp8/common/x86/ |
postproc_mmx.asm | 67 movq mm6, [rbx + 32 ] ; mm6 = kernel 2 taps 71 pmullw mm3, mm6 ; mm3 *= kernel 2 modifiers 73 movq mm6, [rbx + 48] ; mm6 = kernel 3 taps 76 pmullw mm6, mm5 ; mm6 *= p0..p3 * kernel 3 modifiers 77 paddusw mm3, mm6 ; mm3 += mm6 86 movq mm6, [rbx + 64 ] ; mm6 = kernel 4 modifier [all...] |
postproc_mmx.c | 94 movq mm6, [ebx + 32 ] ; 95 mm6 = kernel 2 taps 102 pmullw mm3, mm6 ; 105 movq mm6, [ebx + 48] ; 106 mm6 = kernel 3 taps 111 pmullw mm6, mm5 ; 112 mm6 *= p0..p3 * kernel 3 modifiers 113 paddusw mm3, mm6 ; 114 mm3 += mm6 128 movq mm6, [ebx + 64 ] local 141 movq mm6, mm1 ; local 150 por mm7, mm6 ; local 155 movq mm6, [ebx ] ; local 168 movq mm6, mm1 ; local 177 por mm7, mm6 ; local 180 movq mm6, [ebx + 16] ; local 193 movq mm6, mm1 ; local 202 por mm7, mm6 ; local 280 movq mm6, [ebx + 64 ] local 294 movq mm6, mm1 ; local 303 por mm7, mm6 ; local 307 movq mm6, [ebx ] local 321 movq mm6, mm1 ; local 330 por mm7, mm6 ; local 333 movq mm6, [ebx + 16] local 345 movq mm6, mm1 ; local 354 por mm7, mm6 ; local 741 pxor mm6, mm6 ; local 759 paddd mm6, mm1 ; local 784 paddd mm6, mm2 local 791 psubd mm6, mm1 local [all...] |
loopfilter_mmx.asm | 52 movq mm6, mm1 ; q2 54 psubusb mm2, mm6 ; q3-=q2 61 psubusb mm4, mm6 ; q1-=q2 62 psubusb mm6, mm3 ; q2-=q1 63 por mm4, mm6 ; abs(q2-q1) 117 movq mm6, mm5 ; p0 120 psubusb mm3, mm6 ; q0-=p0 157 pxor mm6, [GLOBAL(t80)] ; offset to convert to signed values 160 psubsb mm0, mm6 ; q0 - p0 195 paddsb mm6, mm2 ; p0+= p0 ad [all...] |
iwalsh_mmx.asm | 81 movq mm6, mm4 ;temp al 84 psubw mm6, mm5 ;al - bl 103 movq mm1, mm6 ; 23 22 21 20 104 punpcklwd mm6, mm5 ; 31 21 30 20 110 punpckldq mm0, mm6 ; 30 20 10 00 aka ip[0] 111 punpckhdq mm4, mm6 ; 31 21 11 01 aka ip[4] 122 movq mm6, mm1 ;temp al 125 psubw mm6, mm5 ;al - bl 139 movq mm4, mm6 ; 23 22 21 20 140 punpcklwd mm6, mm5 ; 31 21 30 2 [all...] |
subpixel_mmx.asm | 44 movq mm6, [rdx + 48] ; 74 pmullw mm4, mm6 ; mm5 *= kernel 3 modifiers 142 movq mm6, [rbx + 48] ; 175 pmullw mm4, mm6 ; mm4 *= kernel 3 modifiers. 264 movq mm6, mm5 ; 267 punpckhbw mm6, mm0 ; 270 pmullw mm6, mm2 ; 273 paddw mm4, mm6 ; 296 movq mm6, mm5 ; 299 punpckhbw mm6, mm0 [all...] |
idctllm_mmx.asm | 80 movq mm6, mm2 ; a1 88 psubw mm6, mm3 ;3 96 punpcklwd mm3, mm6 ; 31 21 30 20 97 punpckhwd mm4, mm6 ; 33 23 32 22 138 movq mm6, mm2 ; a1 146 psubw mm6, mm3 ;3 152 psraw mm6, 3 160 punpcklwd mm3, mm6 ; 31 21 30 20 161 punpckhwd mm4, mm6 ; 33 23 32 22
|
/external/libvpx/vp8/encoder/x86/ |
sad_mmx.asm | 44 pxor mm6, mm6 69 punpcklbw mm0, mm6 70 punpcklbw mm2, mm6 72 punpckhbw mm1, mm6 73 punpckhbw mm3, mm6 91 punpcklwd mm0, mm6 92 punpckhwd mm7, mm6 136 pxor mm6, mm6 [all...] |
quantize_mmx.asm | 83 movq mm6, mm7 84 pcmpgtw mm5, mm6 86 pandn mm5, mm6 90 movq mm6, [rcx+8] 92 paddw mm7, mm6 104 movq mm6, [rax+8] 106 pmullw mm7, mm6 124 movq mm6, mm7 125 pcmpgtw mm5, mm6 127 pandn mm5, mm6 [all...] |
variance_impl_mmx.asm | 88 pxor mm6, mm6 ; Blank mmx7 102 punpcklbw mm0, mm6 ; unpack to higher prrcision 103 punpcklbw mm1, mm6 104 punpckhbw mm2, mm6 ; unpack to higher prrcision 105 punpckhbw mm3, mm6 126 punpcklbw mm0, mm6 ; unpack to higher prrcision 127 punpcklbw mm1, mm6 128 punpckhbw mm2, mm6 ; unpack to higher prrcision 129 punpckhbw mm3, mm6 [all...] |
encodeopt.asm | 82 movq mm6, [rdi+8] 86 psubw mm5, mm6 103 movq mm6, [rdi+24] 104 psubw mm5, mm6 156 movq mm6, [rdi+8] 159 psubw mm5, mm6 174 movq mm6, [rdi+24] 175 psubw mm5, mm6
|
sad_sse3.asm | 206 movq mm6, QWORD PTR [%3+1] 210 psadbw mm6, mm0 223 paddw mm6, mm2 241 paddw mm6, mm2 316 movq mm6, QWORD PTR [%5] 321 psadbw mm6, mm0 336 paddw mm6, mm3 353 paddw mm6, mm3 471 punpckldq mm5, mm6 496 punpckldq mm5, mm6 [all...] |
/external/llvm/test/CodeGen/X86/ |
2008-02-26-AsmDirectMemOp.ll | 15 tail call void asm sideeffect "movd $1, %mm6 \0A\09packssdw %mm6, %mm6 \0A\09packssdw %mm6, %mm6 \0A\09movd $2, %mm5 \0A\09pxor %mm7, %mm7 \0A\09packssdw %mm5, %mm5 \0A\09packssdw %mm5, %mm5 \0A\09psubw %mm5, %mm7 \0A\09pxor %mm4, %mm4 \0A\09.align 1<<4\0A\091: \0A\09movq ($0, $3), %mm0 \0A\09movq 8($0, $3), %mm1 \0A\09pmullw %mm6, %mm0 \0A\09pmullw %mm6, %mm1 \0A\09movq ($0, $3), %mm2 \0A\09movq 8($0, $3), %mm3 \0A\09pcmpgtw %mm4, %mm2 \0A\09pcmpgtw %mm4, %mm3 \0A\09pxor %mm2, %mm0 \0A\09pxor %mm3, %mm1 \0A\09paddw %mm7, %mm0 \0A\09paddw %mm7, %mm1 \0A\09pxor %mm0, %mm2 \0A\09pxor %mm1, %mm3 \0A\09pcmpeqw %mm7, %mm0 \0A\09pcmpeqw %mm7, %mm1 \0A\09pandn %mm2, %mm0 \0A\09pandn %mm3, %mm1 \0A\09movq %mm0, ($0, $3) \0A\09movq %mm1, 8($0, $3) \0A\09add $$16, $3 \0A\09jng 1b \0A\09", "r,imr,imr,r,~{dirflag},~{fpsr},~{flags},~{memory}"( i16* null, i32 %tmp1, i32 0, i32 0 ) nounwind
|
/external/libvpx/vpx_scale/win32/ |
scaleopt.c | 84 movq mm6, const35_2 // mm6 = 9a xx 33 xx cd xx 66 xx local 110 pmullw mm1, mm6 // 155 pmullw mm1, mm6 // 215 movq mm6, const45_2 // mm6 = cd xx 9a xx 66 xx 33 xx local 234 pmullw mm1, mm6 // 01*205 02*154 03*102 04* 51 241 pmullw mm3, mm6 // 05*205 06*154 07*102 08* 51 285 pmullw mm1, mm6 // 01*205 02*154 03*102 04* 51 292 pmullw mm3, mm6 // 05*205 06*154 07*102 07* 5 369 movq mm6, four_fifths \/\/ constan local 400 movq mm6, three_fifths local 444 movq mm6, mm3 \/\/ make a copy local 445 pmullw mm6, mm5 \/\/ d * 2\/5 local 467 movq mm6, one_fifth \/\/ mm6 = 1\/5 local 553 movq mm6, four_fifths \/\/ constan local 584 movq mm6, three_fifths local 630 movq mm6, mm3 \/\/ make a copy local 631 pmullw mm6, mm5 \/\/ d * 2\/5 local 710 movq mm6, three_fifths \/\/ mm6 = 3\/5 local 749 movq mm6, mm0 \/\/ make copy of c low local 752 pmullw mm6, one_fifth \/\/ c * 1\/5 low local 759 movq mm6, mm0 \/\/ make copy of c low local 761 pmullw mm6, four_fifths \/\/ c * 4\/5 low local 795 movq mm6, two_fifths \/\/ mm6 = 2\/5 local 882 movq mm6, three_fifths \/\/ mm6 = 3\/5 local 925 movq mm6, mm0 \/\/ make copy of c low local 928 pmullw mm6, one_fifth \/\/ c * 1\/5 low local 935 movq mm6, mm0 \/\/ make copy of c low local 937 pmullw mm6, four_fifths \/\/ c * 4\/5 low local 1018 movq mm6, four_ones \/\/ mm6= 1, 1, 1, 1 local 1125 movq mm6, four_ones local 1286 movq mm6, const54_2 ; local 1370 movd mm6, [eax+ecx*2] local 1378 punpcklbw mm6, mm7 local 1381 pmullw mm6, three_fourths local 1438 movq mm6, const53_2 ; local 1533 movq mm6, two_thirds local 1678 movq mm6, round_values; local [all...] |
/external/libvpx/vp8/decoder/x86/ |
dequantize_mmx.asm | 120 movq mm6, mm2 ; a1 128 psubw mm6, mm3 ;3 136 punpcklwd mm3, mm6 ; 31 21 30 20 137 punpckhwd mm4, mm6 ; 33 23 32 22 178 movq mm6, mm2 ; a1 186 psubw mm6, mm3 ;3 192 psraw mm6, 3 200 punpcklwd mm3, mm6 ; 31 21 30 20 201 punpckhwd mm4, mm6 ; 33 23 32 22 322 movq mm6, mm2 ; a [all...] |
/external/flac/libFLAC/ia32/ |
fixed_asm.nasm | 130 movq mm6, mm7 ; mm6 = 0:error_0 132 punpckldq mm6, mm7 ; mm6 = error_1:error_0 133 movq mm5, mm6 ; mm5 = error_1:error_0 134 movq mm7, mm6 ; mm7 = error_1:error_0 136 movq mm3, mm6 ; mm3 = error_1:error_0 137 psrad mm6, 31 138 pxor mm7, mm6 139 psubd mm7, mm6 ; mm7 = abs(error_1):abs(error_0 [all...] |
/external/qemu/distrib/sdl-1.2.12/src/audio/ |
SDL_mixer_MMX_VC.c | 83 movq mm6, mm4 //%%mm4,%%mm6\n" // mm6 = l(e*v)|l(f*v)|l(g*v)|l(h*v) local 86 punpcklwd mm6, mm5 //%%mm5,%%mm6\n" // mm6 = g*v|h*v local 96 psrad mm6, 7 //$7,%%mm6\n" // mm6= g*v/128|h*v/128 local 99 packssdw mm6, mm4 //%%mm4,%%mm6\n" // mm6 = s(e*v|f*v|g*v|h*v local 101 paddsw mm6, mm5 \/\/%%mm5,%%mm6\\n" \/\/ mm6 = adjust_volume(src)+dst local [all...] |
SDL_mixer_MMX.c | 87 " movq %%mm4,%%mm6\n" /* mm6 = l(e*v)|l(f*v)|l(g*v)|l(h*v) */ 92 " punpcklwd %%mm5,%%mm6\n" /* mm6 = g*v|h*v */ 108 " psrad $7,%%mm6\n" /* mm6= g*v/128|h*v/128 */ 112 " packssdw %%mm4,%%mm6\n" /* mm6 = s(e*v|f*v|g*v|h*v) */ 115 " paddsw %%mm5,%%mm6\n" /* mm6 = adjust_volume(src)+dst * [all...] |
/external/qemu/distrib/sdl-1.2.12/src/video/ |
SDL_yuv_mmx.c | 133 // create R (result in mm6) 135 "movq %%mm4,%%mm6\n" // 0 L2 0 L0 0 l2 0 l0 137 "paddsw %%mm1, %%mm6\n" // lum1+red:x R2 x R0 x r2 x r0 139 "packuswb %%mm6,%%mm6\n" // R2 R0 r2 r0 R2 R0 r2 r0 141 "punpcklbw %%mm5,%%mm6\n" // R3 R2 R1 R0 r3 r2 r1 r0 175 // fill destination row1 (needed are mm6=Rr,mm7=Gg,mm5=Bb) 179 "movq %%mm6,%%mm1\n" // R3 R2 R1 R0 r3 r2 r1 r0 202 // fill destination row2 (needed are mm6=Rr,mm7=Gg,mm5=Bb) 205 "punpckhbw %%mm2,%%mm6\n" // 0 R3 0 R2 0 R1 0 R [all...] |
SDL_blit_A.c | 254 movq_r2r(mm2, mm6); /* 2 x dst -> mm6(ARGBARGB) */ 259 pand_r2r(mm4, mm6); /* dst & mask -> mm6 */ 261 paddd_r2r(mm6, mm5); /* mm6 + mm5 -> mm5 */ 331 movq_r2r(mm2, mm6); /* 2 x dst -> mm6(ARGBARGB) */ 336 punpckhbw_r2r(mm5, mm6); /* high - 0A0R0G0B -> mm6(dst2) * [all...] |
SDL_RLEaccel.c | 169 movq_r2r(mm2, mm6); \ 173 punpckhbw_r2r(mm6, mm6); \ 177 pand_r2r(mm3, mm6); \ 178 psubw_r2r(mm6, mm1); \ 183 paddw_r2r(mm1, mm6); \ 185 pand_r2r(mm3, mm6); \ 187 packuswb_r2r(mm6, mm6); \ 189 psllq_i2r(32, mm6); \ [all...] |
/external/qemu/distrib/libpng-1.2.19/ |
pngvcrd.c | 171 psubb mm6,mm6 //zero mm6 local 184 pcmpeqb mm0,mm6 185 pcmpeqb mm1,mm6 186 pcmpeqb mm2,mm6 197 movq mm6,mm0 local 199 pandn mm6,mm7 local 200 por mm4,mm6 207 movq mm6,[ebx+8 local 212 movq mm6,[esi+16] local 213 pand mm6,mm2 local 217 por mm6,mm4 local 307 psubb mm6,mm6 \/\/zero mm6 local 337 movq mm6,mm0 local 339 pandn mm6,mm7 local 346 movq mm6,[ebx+8] local 351 movq mm6,[esi+16] local 352 pand mm6,mm2 local 356 por mm6,mm4 local 449 psubb mm6,mm6 \/\/zero mm6 local 468 movq mm6,mm0 local 469 pandn mm6,[ebx] local 719 psubb mm6,mm6 \/\/zero mm6 local 742 movq mm6,mm0 local 744 pandn mm6,mm7 local 751 movq mm6,[ebx+8] local 840 psubb mm6,mm6 \/\/zero mm6 local 876 movq mm6,mm0 local 877 pandn mm6,[ebx] local 881 movq mm6,[esi+8] local 882 pand mm6,mm1 local 885 por mm6,mm7 local 888 movq mm6,[esi+16] local 889 pand mm6,mm2 local 892 por mm6,mm7 local 897 movq mm6,mm3 local 898 pandn mm6,[ebx+24] local 902 movq mm6,[esi+32] local 903 pand mm6,mm4 local 906 por mm6,mm7 local 911 movq mm6,mm5 local 912 pandn mm6,[ebx+40] local 1252 movq mm6, mm0 ; X X v2 v1 v0 v5 v4 v3 local 1255 psrlq mm6, 24 ; 0 0 0 X X v2 v1 v0 local 1257 movq mm5, mm6 ; 0 0 0 X X v2 v1 v0 local 1258 psllq mm6, 8 ; 0 0 X X v2 v1 v0 0 local 1262 por mm6, mm5 ; 0 0 X X v2 v1 v0 v2 local 1263 movd [edi+8], mm6 ; move double to memory local 2026 movq mm6, mm7 local 2042 psllq mm6, ShiftBpp \/\/ shift the mm6 mask to cover bytes 3-5 local 2056 psllq mm6, ShiftBpp \/\/ shift the mm6 mask to cover the last two local 2101 movq mm6, mm7 local 2103 psllq mm6, ShiftBpp \/\/ Create mask for 2nd active group local 2175 movq mm6, mm7 local 2187 psllq mm6, ShiftBpp \/\/ shift the mm6 mask to cover bytes 2 & 3 local 2200 psllq mm6, ShiftBpp \/\/ shift the mm6 mask to cover bytes 4 & 5 local 2215 psllq mm6, ShiftBpp \/\/ shift the mm6 mask to cover bytes 6 & 7 local 2519 movq mm6, mm4 local 2526 paddw mm6, mm5 local 2537 psubw mm6, mm0 local 2540 psubw mm6, mm0 local 2579 movq mm6, mm5 local 2580 paddw mm6, mm4 local 2596 psubw mm6, mm0 local 2599 psubw mm6, mm0 local 2639 movq mm6, mm4 local 2642 paddw mm6, mm5 local 2658 psubw mm6, mm0 local 2661 psubw mm6, mm0 local 2728 movq mm6, mm4 local 2734 paddw mm6, mm5 local 2745 psubw mm6, mm0 local 2748 psubw mm6, mm0 local 2772 movq mm6, mm2 local 2775 psllq mm6, ShiftBpp local 2792 movq mm6, mm4 local 2798 paddw mm6, mm5 local 2809 psubw mm6, mm0 local 2812 psubw mm6, mm0 local 2868 movq mm6, mm4 local 2874 paddw mm6, mm5 local 2885 psubw mm6, mm0 local 2888 psubw mm6, mm0 local 2924 movq mm6, mm4 local 2930 paddw mm6, mm5 local 2941 psubw mm6, mm0 local 2944 psubw mm6, mm0 local 2999 movq mm6, mm4 local 3005 paddw mm6, mm5 local 3016 psubw mm6, mm0 local 3019 psubw mm6, mm0 local 3056 movq mm6, mm4 local 3062 paddw mm6, mm5 local 3073 psubw mm6, mm0 local 3076 psubw mm6, mm0 local 3329 movq mm6, mm7 local 3331 psllq mm6, ShiftBpp \/\/ Move mask in mm6 to cover 3rd active local 3436 movq mm6, mm7 local 3438 psllq mm6, ShiftBpp \/\/ Move mask in mm6 to cover 3rd active local 3511 movq mm6, [edi+ebx+48] \/\/ Load Sub(x) for 7th 8 bytes local 3513 paddb mm6, mm5 local 3625 movq mm6, [edi+ebx+24] local 3627 paddb mm6, mm7 local 3641 movq mm6, [edi+ebx+56] local 3644 paddb mm6, mm7 local [all...] |
/external/qemu/distrib/sdl-1.2.12/src/hermes/ |
mmxp2_32.asm | 65 ; set up mm6 as the mask, mm7 as zero 66 load_immq mm6, mmx32_rgb888_mask 78 pand mm0, mm6 ; 0 R G B 0 r g b 80 pand mm1, mm6 ; 0 R G B 0 r g b 130 load_immq mm6, mmx32_rgb565_g 142 pand mm0, mm6 ; 00g0 152 pand mm4, mm6 ; 00g0 197 load_immq mm6, mmx32_rgb565_g 209 pand mm0, mm6 ; 0 0 g 0 221 pand mm4, mm6 ; 0 0 g [all...] |
/external/bluetooth/bluez/sbc/ |
sbc_primitives_mmx.c | 130 "movq 48(%0), %%mm6\n" 134 "pmaddwd 48(%1), %%mm6\n" 138 "paddd %%mm6, %%mm2\n" 143 "movq 80(%0), %%mm6\n" 147 "pmaddwd 80(%1), %%mm6\n" 151 "paddd %%mm6, %%mm2\n" 156 "movq 112(%0), %%mm6\n" 160 "pmaddwd 112(%1), %%mm6\n" 164 "paddd %%mm6, %%mm2\n" 169 "movq 144(%0), %%mm6\n [all...] |
/packages/apps/Camera/jni/feature_stab/db_vlvm/ |
db_feature_detection.cpp | 107 pxor mm6,mm6 /*2 Set to zero*/ local 117 pcmpgtb mm6,mm0 /*10 Create unpack mask*/ local 119 punpcklbw mm0,mm6 /*11 Unpack low*/ 120 punpckhbw mm1,mm6 /*12 Unpack high*/ 139 pxor mm6,mm6 /*29 Set to zero*/ local 141 pcmpgtw mm6,mm0 /*31 Create unpack mask for dx*dx*/ local 142 punpcklwd mm0,mm6 /*32 Unpack dx*dx lows*/ 144 punpckhwd mm5,mm6 /*33 Unpack dx*dx highs* 145 pxor mm6,mm6 \/*36 Set to zero*\/ local 149 pcmpgtw mm6,mm4 \/*38 Create unpack mask for dx*dy*\/ local 153 pxor mm6,mm6 \/*43 Set to zero*\/ local 157 pcmpgtw mm6,mm2 \/*45 Create unpack mask for dy*dy*\/ local 167 pxor mm6,mm6 \/*54 Set to zero*\/ local 169 pcmpgtw mm6,mm1 \/*56 Create unpack mask for dx*dx*\/ local 180 pxor mm6,mm6 \/*68 Set to zero*\/ local 183 pcmpgtw mm6,mm3 \/*70 Create unpack mask for dy*dy*\/ local 301 movq mm6,[ecx+512] \/*19 Get dxy2*\/ local 305 pslld mm6,1 \/*20 Shift dxy2 1*\/ local 307 pslld mm6,1 \/*22 Shift dxy2 2*\/ local 399 movq mm6,[eax] \/*1 Load new data*\/ local 402 pslld mm6,1 \/*4*\/ local 404 pslld mm6,1 \/*6*\/ local 406 pshufw mm6,mm6,4Eh \/*8 Swap the two double-words using bitmask 01001110=4Eh*\/ local 409 pand mm6,mm7 \/*11 Get low double-word only*\/ local 411 pxor mm6,mm5 \/*13 Get high double-word only*\/ local 436 movq mm6,[eax+16] \/*33 Load new data*\/ local 441 pslld mm6,1 \/*36*\/ local 443 pslld mm6,1 \/*38*\/ local 445 pshufw mm6,mm6,4Eh \/*40 Swap the two double-words using bitmask 01001110=4Eh*\/ local 449 pand mm6,mm7 \/*43 Get low double-word only*\/ local 451 pxor mm6,mm5 \/*45 Get high double-word only*\/ local 453 movq mm6,mm4 \/*50a Copy*\/ local [all...] |
/packages/apps/LegacyCamera/jni/feature_stab/db_vlvm/ |
db_feature_detection.cpp | 107 pxor mm6,mm6 /*2 Set to zero*/ local 117 pcmpgtb mm6,mm0 /*10 Create unpack mask*/ local 119 punpcklbw mm0,mm6 /*11 Unpack low*/ 120 punpckhbw mm1,mm6 /*12 Unpack high*/ 139 pxor mm6,mm6 /*29 Set to zero*/ local 141 pcmpgtw mm6,mm0 /*31 Create unpack mask for dx*dx*/ local 142 punpcklwd mm0,mm6 /*32 Unpack dx*dx lows*/ 144 punpckhwd mm5,mm6 /*33 Unpack dx*dx highs* 145 pxor mm6,mm6 \/*36 Set to zero*\/ local 149 pcmpgtw mm6,mm4 \/*38 Create unpack mask for dx*dy*\/ local 153 pxor mm6,mm6 \/*43 Set to zero*\/ local 157 pcmpgtw mm6,mm2 \/*45 Create unpack mask for dy*dy*\/ local 167 pxor mm6,mm6 \/*54 Set to zero*\/ local 169 pcmpgtw mm6,mm1 \/*56 Create unpack mask for dx*dx*\/ local 180 pxor mm6,mm6 \/*68 Set to zero*\/ local 183 pcmpgtw mm6,mm3 \/*70 Create unpack mask for dy*dy*\/ local 301 movq mm6,[ecx+512] \/*19 Get dxy2*\/ local 305 pslld mm6,1 \/*20 Shift dxy2 1*\/ local 307 pslld mm6,1 \/*22 Shift dxy2 2*\/ local 399 movq mm6,[eax] \/*1 Load new data*\/ local 402 pslld mm6,1 \/*4*\/ local 404 pslld mm6,1 \/*6*\/ local 406 pshufw mm6,mm6,4Eh \/*8 Swap the two double-words using bitmask 01001110=4Eh*\/ local 409 pand mm6,mm7 \/*11 Get low double-word only*\/ local 411 pxor mm6,mm5 \/*13 Get high double-word only*\/ local 436 movq mm6,[eax+16] \/*33 Load new data*\/ local 441 pslld mm6,1 \/*36*\/ local 443 pslld mm6,1 \/*38*\/ local 445 pshufw mm6,mm6,4Eh \/*40 Swap the two double-words using bitmask 01001110=4Eh*\/ local 449 pand mm6,mm7 \/*43 Get low double-word only*\/ local 451 pxor mm6,mm5 \/*45 Get high double-word only*\/ local 453 movq mm6,mm4 \/*50a Copy*\/ local [all...] |